Networks of interacting genes control a variety of fundamental biological processes, including embryonic development, cell division, and cell death. Understanding the function of an entire network of genes which control development is a central problem of functional genomics and systems biology. Although general technology to solve the problem does not yet exist, it can be solved today for the network of genes which control the segmentation system of Drosophila melanogaster. The research plan to accomplish this goal is an integrated program of experimentation and computational modeling, which will be used to characterize the process by which positional information encoded in maternal gradients is transformed into precise and stable expression of the segment polarity genes engrailed and wingless in stripes one nucleus in width. This process creates an extremely precise spatial body plan from imprecise maternal cues. Early patterns containing considerable noise and variation between individuals are transformed into highly precise late patterns with little variation. The work proposed will result in a realistic and predictive theory of pattern formation and error correction in the morphogenetic field that determines Drosophila segmentation. The specific aims are to 1) Make use of a chemical kinetic model describing the the evolution of the expression pattern of the segmentation genes during the blastoderm stage to precisely understand how each expression domain of every segmentation gene forms where and when it does. 2) Obtain a new set of quantitative data at both the RNA and protein levels in both wild type and mutants. We will use this new data to 3) Understand and characterize how variation between embryos is canalized into a reliable and size compensated pattern in both wild type and mutants. Accomplishing these aims will require us to 4)Construct a prototype Laboratory Information Management System (LIMS) with new simulation tools connected to an expanded internet database to make all of our data and simulation tools available on the web.